A virtual array based antenna direction finding method, apparatus and device

Abstract

The application discloses a virtual array-based antenna direction finding method, device and equipment. When performing a direction finding task of a target antenna for the first time, a moving track of the target antenna can be determined according to a direction finding requirement, and a plurality of physical sampling positions can be selected uniformly on the track M The three-dimensional coordinates of the physical sampling positions are recorded. Then the physical sampling positions are equivalent to virtual array antenna elements, and a virtual array antenna is formed according to the track arrangement. A narrowband signal source is received on the virtual array to convert into a complex signal matrix, and a DBF angle estimation is performed on the virtual array antenna based on the complex signal matrix to obtain a preliminary DOA estimation angle, so as to construct adjacent and difference beams based on the estimation angle. The estimation deviation is solved by using the adjacent beams, and the preliminary estimation angle is corrected to obtain a final accurate angle estimation result. The application can improve the DOA estimation precision without increasing the hardware cost and system complexity.

Description

Technical Field

[0001] This application relates to the field of communication technology, specifically to an antenna angle measurement method, apparatus, and device based on a virtual array. Background Technology

[0002] In the field of wireless positioning technology, Direction of Arrival (DOA) estimation is a core component for achieving high-precision positioning. Its core objective is to calculate the angle of arrival (i.e., azimuth and elevation) of a signal propagating from the transmitter to the receiver. Traditional DOA estimation methods generally rely on multi-antenna arrays, which consist of multiple independent antenna elements arranged according to a preset structure. Their core spatial characteristic parameter is the array aperture. The array aperture refers to the effective spatial distribution range of the antenna array, determined by the arrangement, number, and spacing of the antenna elements.

[0003] According to array signal processing theory, the angular resolution of an antenna array is positively correlated with its aperture: the larger the aperture, the smaller the minimum angular difference the system can resolve, and the higher the DOA estimation accuracy and positioning accuracy. However, in practical applications, improving positioning accuracy by increasing the array aperture faces several technical bottlenecks: First, increasing the array aperture requires a simultaneous increase in the number of antenna elements, each of which needs to be configured with an independent RF link, analog-to-digital conversion module, and signal processing unit, leading to a significant increase in hardware costs and system power consumption. Second, the physical size of the array increases with the aperture, making it difficult to adapt to space-constrained applications such as mobile terminals, drones, and robots. Third, traditional antenna arrays have a fixed configuration and cannot dynamically adjust their spatial positions according to different measurement requirements, limiting the efficient use of limited space resources. Fourth, the antenna element density increases with the aperture, significantly enhancing the electromagnetic coupling effect between elements, which in turn affects the accuracy of signal phase measurement, leading to a decrease in DOA estimation accuracy.

[0004] Therefore, how to improve the accuracy of DOA estimation while controlling hardware costs and system complexity is a technical problem that urgently needs to be solved by those skilled in the art. Summary of the Invention

[0005] To address the aforementioned issues, this application provides an antenna angle measurement method, apparatus, and device based on a virtual array, which can improve DOA estimation accuracy while controlling hardware costs and system complexity.

[0006] The embodiments of this application disclose the following technical solutions:

[0007] An antenna angle measurement method based on a virtual array, characterized in that the method includes:

[0008] When performing the angle measurement task of the target antenna for the first time, the movement trajectory of the target antenna is determined according to the application angle measurement requirements of the target antenna, and uniform selections are made on the movement trajectory. M Each physical sampling position is recorded, and the three-dimensional spatial coordinates of each physical sampling position are recorded; the movement trajectory is a straight trajectory adapted to a uniform linear array ULA, or a circular trajectory adapted to a uniform circular array UCA. M It is a positive integer;

[0009] Combining the various three-dimensional spatial coordinates to the M Each physical sampling location corresponds one-to-one with, which is equivalent to M Each virtual array antenna element, and the M The virtual array antenna elements are arranged to match the moving trajectory.

[0010] The virtual array antenna M Wireless signals from a narrowband signal source to be located are collected at each physical sampling location. The wireless signals are then subjected to quadrature demodulation and analog-to-digital conversion using an orthogonal receiver to obtain a complex signal matrix. The wireless signals include signal amplitude, phase, and additive white Gaussian noise.

[0011] The virtual array antenna is subjected to DBF angle estimation using the complex signal matrix to obtain a preliminary DOA estimated angle. Based on the preliminary DOA estimated angle and a preset scanning interval, adjacent sum and difference beams are constructed to obtain angle-measuring adjacent beams. The preliminary DOA estimated angle includes a preliminary azimuth angle and / or a preliminary elevation angle. The angle-measuring adjacent beams include an azimuth sum and difference beam and / or an elevation sum and difference beam. The azimuth sum and difference beam includes an azimuth sum beam and an azimuth difference beam. The elevation sum and difference beam includes an elevation sum beam and an elevation difference beam.

[0012] The angle estimation deviation is calculated based on the adjacent beams of the angle measurement, and the angle estimation deviation is used to correct and compensate the initial angle estimate of the DOA to obtain the final angle estimation result; the angle estimation deviation includes the azimuth angle estimation deviation and / or the elevation angle estimation deviation.

[0013] An antenna angle measurement device based on a virtual array, the device comprising:

[0014] The selection unit, when performing the angle measurement task of the target antenna for the first time, is used to determine the movement trajectory of the target antenna according to the application angle measurement requirements of the target antenna, and uniformly selects along the movement trajectory. M Each physical sampling position is recorded, along with its three-dimensional spatial coordinates. The movement trajectory is either a straight line trajectory adapted to ULA or a circular trajectory adapted to UCA. M It is a positive integer;

[0015] Virtual array antenna construction unit, used to combine various three-dimensional spatial coordinates to... M Each physical sampling location corresponds one-to-one with, which is equivalent to M Each virtual array antenna element, and the M The virtual array antenna elements are arranged to match the moving trajectory.

[0016] Complex signal matrix construction unit, used in the virtual array antenna M Wireless signals from a narrowband signal source to be located are collected at each physical sampling location. The wireless signals are then subjected to quadrature demodulation and analog-to-digital conversion using an orthogonal receiver to obtain a complex signal matrix. The wireless signals include signal amplitude, phase, and additive white Gaussian noise.

[0017] The preliminary estimation unit is used to perform DBF angle estimation on the virtual array antenna in conjunction with the complex signal matrix to obtain the preliminary DOA estimated angle; the preliminary DOA estimated angle includes a preliminary azimuth angle and / or a preliminary elevation angle.

[0018] An angle-measuring adjacent beam construction unit is used to construct adjacent sum and difference beams based on the initial angle estimate of the DOA and the preset scanning interval to obtain angle-measuring adjacent beams; the angle-measuring adjacent beams include azimuth sum and difference beams and / or elevation sum and difference beams; the azimuth sum and difference beams include azimuth sum beams and azimuth difference beams; the elevation sum and difference beams include elevation sum beams and elevation difference beams;

[0019] The correction and compensation unit is used to solve the angle estimation deviation based on the adjacent beams of the angle measurement, and to use the angle estimation deviation to correct and compensate the initial estimated angle of the DOA to obtain the final angle estimation result; the angle estimation deviation includes the azimuth angle estimation deviation and / or the elevation angle estimation deviation.

[0020] An antenna angle measurement device based on a virtual array includes: a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, it implements the antenna angle measurement method based on the virtual array as described above.

[0021] Compared with the prior art, this application has the following beneficial effects:

[0022] This application provides an antenna angle measurement method, apparatus, and device based on a virtual array. Specifically, during the initial execution of the target antenna angle measurement task, a single target antenna is controlled to move along a preset trajectory and select... MEach physical sampling location is equivalent to a virtual array antenna element. Without increasing the number of physical antennas or adding extra RF links, a virtual large-aperture array matching the moving trajectory can be constructed. This significantly reduces hardware costs and system power consumption while avoiding the mutual coupling effect caused by the dense element density of traditional fixed arrays. Furthermore, the configuration of the virtual array can flexibly adapt to one-dimensional and two-dimensional angle measurement requirements by choosing either a linear trajectory of a Uniform Linear Array (ULA) or a circular trajectory of a Uniform Circular Array (UCA), overcoming the limitation of fixed arrays' inability to dynamically adjust and efficiently utilizing spatial resources. In addition, a two-stage processing logic of "Digital Beamforming (DBF) coarse estimation + sum-difference beam fine estimation" is adopted. First, digital beamforming quickly locks the DOA for preliminary angle estimation. Then, based on a preset scanning interval, sum-difference beams in the azimuth and elevation dimensions are constructed to solve for the deviation. This eliminates the need for complex matrix eigenvalue decomposition, significantly improving angle measurement accuracy while reducing system computational complexity. Attached Figure Description

[0023] To more clearly illustrate the technical solutions in this embodiment or the prior art, the drawings used in the description of the embodiment or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0024] Figure 1 A flowchart illustrating an antenna angle measurement method based on a virtual array, provided in this application embodiment;

[0025] Figure 2 A flowchart illustrating another antenna angle measurement method based on a virtual array provided in this application embodiment;

[0026] Figure 3 A flowchart of a DOA angle estimation method provided in this application embodiment;

[0027] Figure 4 A flowchart illustrating an adjacent sum-difference beamforming method provided in this application embodiment;

[0028] Figure 5 This is a schematic diagram of an antenna angle measuring device based on a virtual array, provided as an embodiment of this application. Detailed Implementation

[0029] To facilitate understanding of the technical solutions provided in the embodiments of this application, the background technology involved in the embodiments of this application will be described below.

[0030] In wireless positioning systems, angle estimation is one of the key technologies for achieving high-precision positioning. Its core objective is to calculate the DOA of the signal, which is the angle of arrival of the signal from the transmitter to the receiver.

[0031] Traditional angle estimation methods primarily rely on antenna arrays, which consist of multiple independent antenna elements arranged in a specific structure. The array aperture, as a core spatial characteristic parameter of the antenna array, refers to the effective size or distribution range of the antenna array in space, and this parameter is determined by the arrangement, number, and spacing of the antenna elements. The core principle of these traditional angle estimation methods is to estimate and determine the direction of arrival (DOA) of the signal by measuring the phase difference generated when the same signal arrives at different antenna elements in the array.

[0032] According to the theory of array signal processing, the angular resolution of an antenna array is proportional to its array aperture: the larger the array aperture of the antenna array, the smaller the minimum angular difference that can be resolved, and the higher the corresponding positioning accuracy.

[0033] However, in practical applications, increasing the array aperture of an antenna array presents several real-world problems: Increasing the aperture requires increasing the number of antenna elements, and each element needs a complete RF link, directly leading to a significant increase in hardware costs and system power consumption. Simultaneously, the physical size of the array increases with the aperture, making it difficult to adapt to space-constrained applications such as mobile devices and drones. Furthermore, fixed arrays cannot dynamically adjust their position to suit different measurement needs, limiting the full utilization of limited space resources. In addition, increasing the aperture results in a denser arrangement of antenna elements, significantly enhancing the mutual coupling effect between elements and thus affecting the accuracy of angle estimation.

[0034] To address this issue, this application provides an antenna angle measurement method, apparatus, and device based on a virtual array. When performing the initial angle measurement task of the target antenna, the target antenna's movement trajectory is first determined according to the actual angle measurement requirements (one-dimensional or two-dimensional): if one-dimensional angle measurement (azimuth or elevation only) is required, a straight trajectory adapted to ULA is selected; if two-dimensional angle measurement (simultaneous estimation of azimuth and elevation) is required, a circular trajectory adapted to UCA is selected. Subsequently, uniform selection is performed on this trajectory. M Each physical sampling location is recorded, and its three-dimensional spatial coordinates are recorded. These sampling locations are then represented as equivalent to... M Each virtual array antenna element is arranged to match the trajectory of the virtual array antenna. Then, in... MWireless signals (including signal amplitude, phase, and additive white Gaussian noise) from the narrowband signal source to be located are collected at each physical sampling location. Orthogonal receivers are used to perform orthogonal demodulation and analog-to-digital conversion on the wireless signals to form a complex signal matrix. Then, a two-stage processing logic of "coarse estimation by digital beamforming (DBF) + fine estimation by sum and difference beams" is employed: first, DBF angle estimation is performed on the virtual array antenna using the complex signal matrix to quickly lock the initial estimated direction of arrival (DOA) angle (including initial azimuth and / or initial elevation angle); then, based on this initial estimated angle and combined with a preset scanning interval, sum and difference beams (including sum and difference beams) are constructed in the azimuth and elevation dimensions respectively, forming angle-measuring adjacent beams. Finally, the angle estimation deviation of the azimuth and / or elevation angles is solved using the angle-measuring adjacent beams, and this deviation is used to correct and compensate for the initial estimated angle, ultimately obtaining a high-precision angle estimation result. This application can improve DOA estimation accuracy without increasing hardware cost or system complexity.

[0035] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of this application.

[0036] See Figure 1 The figure is a flowchart of an antenna angle measurement method based on a virtual array provided in an embodiment of this application. Figure 1 As shown, the antenna angle measurement method based on a virtual array may include steps S101-S105:

[0037] S101: When performing the angle measurement task of the target antenna for the first time, determine the movement trajectory of the target antenna according to the application angle measurement requirements of the target antenna, and uniformly select along the movement trajectory. M Each physical sampling location is recorded, along with its three-dimensional spatial coordinates.

[0038] When performing the angle measurement task of the target antenna for the first time, in order to make the virtual array accurately match the dimensional requirements of the actual angle measurement scenario (one-dimensional only requires estimating the azimuth or elevation angle, while two-dimensional requires estimating both the azimuth and elevation angles simultaneously), the movement trajectory of the target antenna must first be determined according to the application angle measurement requirements of the target antenna: if only one-dimensional angle measurement is required (for example, a drone only needs to determine the horizontal azimuth angle of the signal in front, or a robot only needs to identify the elevation angle of the signal above), then a straight trajectory adapted to ULA is selected. This trajectory allows the virtual array to form an effective aperture in a single dimension, focusing on single-dimensional angle estimation; if two-dimensional angle measurement is required (for example, a base station needs to simultaneously lock the horizontal azimuth and vertical height of a mobile terminal, or a radar needs to accurately locate the azimuth and elevation angles of an aerial target), then a circular trajectory adapted to UCA is selected. Its ring configuration can cover the full angular range in the plane, meeting the requirements of two-dimensional angle measurement.

[0039] After determining the movement trajectory, it is necessary to select evenly on the trajectory. M physical sampling locations ( M It is a positive integer, and its value needs to be combined with the angle measurement accuracy requirements and the hardware sampling capability. M ≥8 (to ensure the aperture effect of the virtual array), the purpose of uniform selection is to ensure that the spacing between virtual array elements is consistent, and to avoid distortion of signal phase difference measurement due to uneven position distribution, which in turn affects the accuracy of angle estimation.

[0040] Simultaneously, the three-dimensional spatial coordinates of each physical sampling location need to be accurately recorded using a positioning module (such as laser positioning or GPS positioning). x , y , z These coordinates (axis coordinates) are the core foundational data for subsequently constructing virtual arrays and calculating signal phase relationships.

[0041] For example, in a one-dimensional angle measurement scenario (UAV horizontal azimuth estimation): the UAV needs to estimate the horizontal azimuth of the signal within a range of -90° to 90° ahead during flight (one-dimensional angle measurement requirement). Therefore, the target antenna's movement trajectory is determined to be a straight line along the UAV's horizontal axis (adapted to ULA). Twelve physical sampling positions are then uniformly selected along this straight line trajectory. M =12), the spacing between adjacent sampling positions is set to 0.5λ (λ is the wavelength of the wireless signal, ensuring effective differentiation of signal phase difference), and the three-dimensional spatial coordinates of each position are recorded by the laser positioning module on the device. For example, the coordinates of the first position are (0,0,1.2), the coordinates of the second position are (0.5λ,0,1.2), ..., the coordinates of the twelfth position are (5.5λ,0,1.2), and so on. y axis, z Keep the axis coordinates consistent, only x The axis coordinates are evenly distributed along a straight line trajectory to ensure that the virtual array focuses on receiving and processing signals in the horizontal azimuth dimension.

[0042] Two-dimensional angle measurement scenario (base station omnidirectional target positioning): The base station needs to cover the signal positioning of the surrounding 360° horizontal range and -30°~30° vertical range (two-dimensional angle measurement requirement), therefore the movement trajectory of the target antenna is determined as the radius. r =2 λ A circular trajectory (adapted to UCA) is drawn, with the plane containing the circular trajectory parallel to the ground. Sixteen physical sampling locations are evenly selected on this circular trajectory. M =16), the central angle interval between adjacent sampling positions is 2π / 16=22.5°, and the three-dimensional spatial coordinates of each position are recorded by the GPS positioning module of the base station, for example, the first m The coordinates of each position are [2 λ cos(( m -1)×22.5°,2 λ sin(( m -1)×22.5°),30] (m=1,2,…,16), z The x-axis coordinates are fixed at 30m (base station installation height), and the x-axis and y-axis coordinates are evenly distributed along a circular trajectory to ensure that the virtual array can receive signals from different azimuth and elevation angles from all directions.

[0043] S102: Combining the various three-dimensional spatial coordinates to... M Each physical sampling location corresponds one-to-one with, which is equivalent to M Each virtual array antenna element, and the M The virtual array antenna elements are arranged to match the moving trajectory.

[0044] To achieve angular resolution equivalent to a large-aperture physical array without increasing the number of physical antennas or RF links (addressing the pain points of high hardware cost and poor spatial adaptability of traditional multi-antenna arrays), it is necessary to... M Virtual array mapping is performed on each physical sampling location: the three-dimensional spatial coordinates of each physical sampling location ( x , y , z Each axis coordinate uniquely corresponds to a virtual array antenna element. The core logic of this operation is to use the spatial diversity generated by antenna movement to convert single antenna sampling at different time points into multi-antenna array reception at the same time point. By accurately recording the spatial coordinates of each physical sampling position, the phase difference information when the signal arrives at different positions is preserved.

[0045] At the same time, in order to ensure that the configuration of the virtual array accurately matches the angle measurement requirements, it is necessary to... MThe virtual array antenna elements are arranged in a manner consistent with their physical movement trajectories: if the physical movement trajectory is a straight line (suitable for one-dimensional angle measurement), the virtual array elements are arranged as ULA; if the physical movement trajectory is a circle (suitable for two-dimensional angle measurement), the virtual array elements are arranged as UCA. This one-to-one correspondence between trajectory and configuration ensures that the aperture distribution of the virtual array meets the requirements of array signal processing theory, providing a suitable array model for subsequent operations such as DBF angle estimation and sum-difference beamforming.

[0046] S103: In the virtual array antenna... M Wireless signals from a narrowband signal source to be located are collected at each physical sampling location. The wireless signals are then subjected to quadrature demodulation and analog-to-digital conversion using an orthogonal receiver to obtain a complex signal matrix.

[0047] To obtain the raw signal data required for subsequent angle estimation, while fully preserving the signal's directional characteristics and channel information, it is necessary to perform [further steps] on the virtual array antenna. M At each physical sampling location, wireless signals transmitted by the narrowband complex signal source to be located are sequentially collected. These signals not only contain core information reflecting the direction of the signal source (signal amplitude and phase), but also are superimposed with additive white Gaussian noise (an unavoidable interference component in real communication scenarios) from the wireless channel. After collection, the wireless signals are quadratured and converted from analog to digital using an orthogonal receiver, converting the analog radio frequency signals into digital complex signals. These signals not only contain core information reflecting the direction of the signal source (signal amplitude and phase), but also are superimposed with additive white Gaussian noise (an unavoidable interference component in real communication scenarios) from the wireless channel. After processing, the received signals from all sampling locations need to be integrated according to a preset format to form a complex signal matrix, providing a unified data source for subsequent DBF, sum and difference beamforming, and other processing.

[0048] It should be noted that the core purpose of signal acquisition is to obtain spatial diversity information of the signal through multi-location sampling: different physical sampling positions correspond to different array elements of the virtual array. When the signal from the same signal source reaches different sampling positions, a phase difference will be generated due to the difference in propagation path. These phase differences are the key basis for subsequent estimation of signal DOA. Completely preserving the signal amplitude and phase information is a prerequisite for ensuring the accurate operation of algorithms such as beamforming and angle calculation. The presence of additive white Gaussian noise requires subsequent algorithms to have a certain anti-interference capability, so that the estimation results are more in line with the actual application scenario.

[0049] S104: Combine the complex signal matrix to perform DBF angle estimation on the virtual array antenna to obtain the preliminary estimated DOA angle, and construct adjacent sum and difference beams based on the preliminary estimated DOA angle and the preset scanning interval to obtain angle measurement adjacent beams.

[0050] To quickly pinpoint the approximate direction of the signal and provide a benchmark for subsequent high-precision correction, the first step is to perform DBF angle estimation on the virtual array antenna using a complex signal matrix. This involves dividing the entire DOA estimation range into several search grids with a preset step size, constructing a corresponding array steering vector (i.e., beam weighting vector) for each grid point, and then weighting and summing this weighting vector with the complex signal matrix to obtain the DBF output signal for each grid point, thereby calculating the spatial spectrum. After searching for the spectral peaks of the spatial spectrum, the angle of the grid point corresponding to the maximum value of the spatial spectrum is the preliminary DOA estimated angle. This angle includes the preliminary azimuth angle (in a one-dimensional angle measurement scenario) or the preliminary azimuth angle and preliminary elevation angle (in a two-dimensional angle measurement scenario). The core function of this operation is to quickly narrow down the range of high-precision angle measurement and avoid the high complexity brought about by full-space fine scanning.

[0051] To extract angle deviation information from the initial estimated angle for accurate correction, adjacent sum and difference beams can be constructed based on this initial estimated angle and a preset scanning interval. Centered on the initial estimated angle, the beams are offset to both sides by a preset scanning interval in the azimuth and elevation dimensions, forming two adjacent beams pointing left and right. Then, by weighted combination of these adjacent beams, sum and difference beams (i.e., adjacent angle measurement beams) are constructed in each dimension. The sum beam enhances the signal strength in the target direction, improving the signal-to-noise ratio; the difference beam captures the signal differences caused by the angle deviation, providing a basis for calculating the deviation.

[0052] It should be noted that for a one-dimensional angle measurement scenario, only a sum and difference beam needs to be constructed in a single dimension (azimuth or elevation), while for a two-dimensional angle measurement scenario, sum and difference beams in both azimuth and elevation dimensions need to be constructed simultaneously, with each dimension containing both sum and difference beams.

[0053] In one possible implementation, the complex signal matrix is:

[0054] ;

[0055] Where y is the complex signal matrix. The complex signal matrix is ​​a complex data vector. express M A complex field of 1 × 1 dimension; The first in m element Indicates the first m The signal received at each physical sampling location; It is additive white Gaussian noise. The first in m noise component It follows a mean of 0 and a variance of . Gaussian distribution; For any narrowband complex signal source pair to be locatedM The transmitted signal vector emitted from each location, , The first in m element , A The signal amplitude; For the target antenna in the 1st m The three-dimensional spatial position vector of the i-th physical sampling location is given by the i-th physical sampling location. m The three-dimensional spatial coordinates of each physical sampling location; Angle of arrival of the wireless signal directional vector, , The azimuth angle of the wireless signal; The elevation angle of the wireless signal is denoted as .

[0056] S105: Based on the angle measurement adjacent beams, solve for the angle estimation deviation, and use the angle estimation deviation to correct and compensate for the initial DOA angle to obtain the final angle estimation result.

[0057] To improve the accuracy of the initial DOA angle estimate to the level required for practical applications, it is necessary to solve for the angle estimation deviation based on adjacent angle measurement beams (azimuth and difference beams, elevation and difference beams). The difference beam can accurately capture the signal difference caused by the deviation between the "true angle and the initial estimated angle," and the ratio of the two (monopulsive difference beam ratio) has a clear linear relationship with the angle deviation. This is the core basis for solving the deviation. By establishing this linear relationship model, the azimuth and / or elevation estimation deviations can be directly calculated. These deviations are then superimposed onto the corresponding initial DOA angle estimate to correct and compensate for the coarse estimation result, ultimately yielding a high-precision angle estimation result.

[0058] Based on the content of S101-S105, when performing the angle measurement task of the target antenna for the first time, the target antenna's movement trajectory is first determined according to the application angle measurement requirements of the target antenna. This determines whether the target antenna's movement trajectory is a straight line trajectory (ULA) or a circular trajectory (UCA), and then uniformly selects along this trajectory. M Each physical sampling location is identified, and its three-dimensional spatial coordinates are recorded. Then, these three-dimensional spatial coordinates are combined with... M Each physical sampling location corresponds one-to-one with, which is equivalent to M Each virtual array antenna element, and will M The virtual array antenna elements are arranged to match the moving trajectory. In the virtual array antenna... MWireless signals from the narrowband signal source to be located are collected at each physical sampling location to obtain a complex signal matrix. The complex signal matrix is ​​then combined with the DBF angle estimation of the virtual array antenna to obtain a preliminary DOA (Directory of Alignment) angle. Based on this estimated angle and a preset scanning interval, adjacent sum-difference beams are constructed to obtain angle-measuring adjacent beams. Finally, the angle estimation deviation is calculated based on the angle-measuring adjacent beams, and this deviation is used to correct and compensate for the preliminary DOA angle to obtain the final angle estimation result. The core logic of this application is "coarse positioning followed by fine correction": the preliminary angle estimation locks in the approximate direction of the signal, while the deviation calculation focuses on fine calibration within the "approximate range," avoiding the high complexity of full-space high-precision scanning. Simultaneously, through the synergistic effect of sum-difference beams, noise interference is effectively suppressed, ensuring the accuracy of the deviation calculation and enabling the final estimation result to meet the requirements of high-precision positioning scenarios.

[0059] See Figure 2 , Figure 2 A flowchart of another antenna angle measurement method based on a virtual array provided in this application embodiment is shown, which can be implemented through steps S201-S204:

[0060] S201: When the target antenna is not performing the angle measurement task for the first time and moves repeatedly along the movement trajectory, the final angle estimation result of the previous time is used as the initial estimated angle of the current DOA.

[0061] S202: Based on the preliminary estimated angle of this DOA and the preset scanning interval, adjacent sum and difference beams are constructed to obtain the adjacent beams for this angle measurement;

[0062] Wherein, the preliminary DOA estimated angle includes the preliminary azimuth angle and / or the preliminary elevation angle; the adjacent beams for this angle measurement include the sum and difference beams for the azimuth angle and / or the sum and difference beams for the elevation angle; the sum and difference beams for this elevation angle include the sum beams for the elevation angle and the difference beams for the elevation angle.

[0063] S203: Calculate the angle estimation deviation based on the adjacent beams of the current angle measurement;

[0064] S204: The initial angle estimate of this DOA is corrected and compensated using the current angle estimation deviation to obtain the final angle estimation result;

[0065] The deviation in the current angle estimation includes the deviation in the current azimuth angle estimation and / or the deviation in the current pitch angle estimation.

[0066] It should be noted that the execution process of steps S202-S204 is similar to that of step S104, which involves "constructing adjacent sum and difference beams based on the preliminary estimated angle of the DOA and the preset scanning interval to obtain the angle measurement adjacent beams", and step S105, and will not be described again here.

[0067] By using steps S201-S204, the DBF coarse estimation process can be skipped. The final result of the previous angle measurement can be directly reused as the initial DOA angle estimate for subsequent fine estimation operations. In scenarios where angle measurement is continuously performed along a determined movement trajectory, the operation process for non-first angle measurement is greatly simplified, the computational complexity is reduced, the real-time performance and efficiency of angle measurement are improved, and the high accuracy of angle estimation is guaranteed.

[0068] See Figure 3 , Figure 3 This is a flowchart of a DOA angle estimation method provided in an embodiment of this application. Accordingly, in step S104, the preliminary estimated DOA angle is obtained by performing DBF angle estimation on the virtual array antenna in combination with the complex signal matrix. Specifically, this can be implemented through steps S301-S305:

[0069] S301: Determine the DOA estimation range of the target antenna based on the installation posture of the target antenna and the spatial distribution range of the narrowband signal source, and determine the division step size.

[0070] To avoid meaningless full-space angle searches, reduce the computational complexity of DBF (Dead-Field Analysis), and ensure that the search range fully covers the possible directions of the signal source, the DOA (Direction of Area) estimation range needs to be determined by combining two core factors: first, the target antenna's mounting attitude (which determines the antenna's effective receiving angle coverage, avoiding invalid searches beyond the antenna's receiving capacity); and second, the spatial distribution range of the narrowband signal source (based on prior scene information, locking the angle range where the signal source is likely to exist, further narrowing the search boundary). Based on this, the step size is determined, and the step size needs to balance "estimation accuracy" and "computational efficiency": a smaller step size results in a more refined angle search and higher initial estimation accuracy, but the number of grids increases, leading to an exponential increase in computational load; a larger step size results in higher computational efficiency, but may miss the true signal direction due to excessively large grid spacing, leading to excessively large initial estimation deviations.

[0071] The core objective of this step is to provide a "precise and efficient search framework" for subsequent gridded searches: by reasonably limiting the DOA estimation range, the computational burden of irrelevant angle intervals can be eliminated. By adapting the partitioning step size to the scene, the computational burden on the system can be minimized while ensuring that the initial estimation accuracy meets the requirements of subsequent sum and difference beam correction, thus ensuring the real-time performance of DBF estimation.

[0072] For example, suppose the current scene is a one-dimensional angle measurement scene of a drone (only the azimuth angle is estimated):

[0073] The target antenna is installed at the nose of the UAV, with an orientation of "horizontal forward, receiving angles covering -90° to 90° (left to right) in front of the nose, and a fixed elevation angle of 0° (horizontal reception only)". Angles outside this range cannot effectively receive signals due to fuselage obstruction or antenna gain attenuation. The known signal source is a ground base station, with its spatial distribution concentrated in the azimuth range of 30° to 60° directly in front of the UAV's flight path. Therefore, considering the antenna's orientation (-90° to 90°) and the signal source distribution (30° to 60°), the DOA estimation range is determined to be azimuth 30° to 60° (excluding invalid intervals of -90° to 30° and 60° to 90°). Considering that subsequent sum and difference beam correction can compensate for a deviation of ±1°, the step size is set to 1°. This ensures that the initial estimation error does not exceed half of the step size (0.5°) and keeps the number of grids to only 31 (60°-30°+1=31), keeping the computational load manageable.

[0074] Assume the current scenario is a two-dimensional angle measurement scenario using a base station (simultaneously estimating azimuth and elevation angles):

[0075] The target antenna is a macro base station antenna, erected at a height of 30m, with an erection orientation of "horizontal coverage of 360° azimuth, vertical reception angle coverage of -30° to 30° (elevation angle, covering ground users downwards, no signal source upwards)". The known signal sources are mobile terminals in surrounding cells, spatially distributed with an azimuth angle of 120° to 180° (east of the base station) and an elevation angle of -15° to 0° (lower-level users). Therefore, the DOA estimation range is determined to be azimuth 120° to 180° and elevation angle of -15° to 0° (to avoid invalid searches of the western azimuth and high-altitude elevation angles). Since the base station has high requirements for real-time angle measurement, and the subsequent sum and difference beam correction accuracy can reach 0.1°, the division step size is set to 0.8°, the number of azimuth angle grids is 76 ((180-120) / 0.8+1=76), the number of elevation angle grids is 19 ((0-(-15)) / 0.8+1≈19), and the total number of grids is only 1444, which is far lower than the number of full-space search. At the same time, the initial estimation error can be controlled within 0.4°, which meets the subsequent correction requirements.

[0076] S302: Based on the partitioning step size, the DOA estimation range is divided into grids to obtain... N A search grid is generated, and the direction vector corresponding to the search grid point of each search grid is obtained.

[0077] To transform the continuous DOA estimation range into discrete, computable angle sampling points, and to provide a clear computational benchmark for subsequent array steering vector construction and DBF beam weighting, the DOA estimation range needs to be meshed based on a preset partitioning step size: the originally continuous angle interval is decomposed into smaller, more uniform segments. N Each grid consists of an independent search grid, with the center of each grid being a search grid point corresponding to a specific angle value (one-dimensional angle measurement is a single azimuth / elevation angle, and two-dimensional angle measurement is a combination of azimuth and elevation angles); then, a corresponding direction vector is generated based on this angle value. ( n =1,2,… N , N (As positive integers), this vector is the core bridge connecting the "angle space" and the "array signal processing space," and is the direction vector. The angle information is transformed into a three-dimensional spatial vector form that can be recognized by array signal processing, providing a foundation for the subsequent construction of beam weighting vectors. N It is a positive integer.

[0078] This step transforms the abstract angle search into a concrete vector computation. The discretized search grid not only adapts to the "discrete operation" characteristics of digital signal processing but also balances search accuracy and computational complexity by controlling the step size; while the direction vector... This unifies the calculation paradigm for different trajectory types (ULA / UCA) and different angle measurement dimensions (one-dimensional / two-dimensional), ensuring that the subsequent construction of array guidance vectors has a unified input standard.

[0079] S303: Construct the array steering vector of each search grid as the first beam weighting vector based on the direction vector corresponding to the search grid point of each search grid.

[0080] To enable the virtual array antenna to perform phase weighting and beam focusing on the direction corresponding to each search grid point, and to transform discrete angle searches into signal processing operations that the array can perform, it is necessary to base the direction vector of each search grid point on the signal. Construct array guide vector And directly use it as the first beam weighting vector. The array steering vector is the core characteristic vector of the virtual array, which is determined by fusing the wavelengths of the wireless signals. , M Three-dimensional spatial position vector of each physical sampling location , ,..., With direction vector The spatial information of the beam is used to assign a phase weight to each physical sampling location that matches the target search direction. This ensures that when a signal is incident from that search direction, the received signals at each sampling location are weighted and superimposed in phase, thus enhancing the signal output in that direction. Conversely, signals from non-target directions will experience destructive interference due to phase mismatch, achieving spatial filtering of the beam. This beam weighting vector... for M A complex column vector of dimensionless numbers has the following mathematical form: Each element corresponds to a phase weighting value at a physical sampling location, perfectly adapting to virtual arrays. M Signal processing requirements for each physical sampling location.

[0081] S304: The first beam weighting vector is used to weight and sum the wireless signals received at each physical sampling position in the received signal to obtain the DBF output.

[0082] To quantify the signal energy intensity in the direction corresponding to each search grid point and filter out the grid where the true direction of the signal source is located, the first beam weighting vector needs to be used. right M The received signals from each physical sampling location are weighted and summed to obtain the DBF output. This can be achieved through the first beam weighting vector. The received signal at each sampling location is assigned a phase weight that matches the target search direction. If the signal happens to be incident from the direction of that grid point, the phases of the signals at each sampling location tend to be consistent after weighting, and the summed signal energy will be enhanced. If the signal comes from other directions, the phases of the weighted signals are disordered, and the summed energy will cancel each other out. This process is essentially a "focused reception" of signals in a specific direction through digital beamforming technology.

[0083] in, The first in z n element , indicating the first n The DBF output signal is weighted by grid points.

[0084] S305: Determine the spatial spectrum based on the DBF output, and perform spectral peak search on the spatial spectrum to obtain the preliminary estimated angle of DOA.

[0085] To transform the complex signal energy characteristics of the DBF output into intuitive and comparable scalar values, thereby accurately locating the approximate direction of the signal source, it is necessary to calculate the spatial spectrum ξ based on the DBF output z, and determine the preliminary estimated angle of DOA through spectral peak search. The DBF output z is in complex form (containing amplitude and phase information), while the first peak in the spatial spectrum ξ... n element Phase interference can be stripped away, retaining only the signal's energy strength. The higher the energy value, the closer the direction of the corresponding search grid point is to the true direction of the signal source. By traversing the spatial spectral values ​​of all search grid points, the location of the "spectral peak" with the largest spectral value is found, and the angle corresponding to this location is the preliminary DOA estimated angle.

[0086] It should be noted that the elements in the spatial spectrum ξ are DBF outputs. The square of the modulus of the corresponding element.

[0087] This process transforms "complex signal operation" into "scalar energy comparison," significantly reducing the complexity of direction determination. At the same time, the uniqueness (or significance) of the energy peak can effectively lock the approximate location of the signal source, providing a reliable reference angle for the subsequent accurate correction of the sum and difference beam.

[0088] Steps S301-S305 first determine the estimation range and step size by combining the antenna mounting attitude and signal source distribution range, effectively reducing invalid search intervals and balancing estimation accuracy and computational load. Then, continuous angles are transformed into discrete grid points and a unified direction vector is generated through gridding, which is adapted to the characteristics of digital signal processing and does not require additional format conversion. Next, an array steering vector is constructed based on the direction vector, signal wavelength, and sampling position coordinates and is directly used as the beam weighting vector, simplifying the processing flow and providing clear physical meaning. Subsequently, phase alignment and energy superposition of the target direction signal and phase cancellation interference of non-target direction signals are achieved through conjugate transpose weighted summation, efficiently completing signal directional focusing. Then, the complex signal form DBF output is converted into a scalar space spectrum, removing phase interference to make the signal intensity comparison more intuitive and providing good robustness to additive white Gaussian noise. Finally, the approximate direction of the signal source is locked through simple spectral peak search, without the need for complex matrix eigenvalue decomposition, resulting in low computational cost. At the same time, it can accurately limit a small range of correction intervals for subsequent sum and difference beam estimation, further reducing the complexity of the overall angle measurement process.

[0089] In one possible implementation, to ensure that the 3D spatial position vector of the physical sampling location and the direction vector of the search grid point are accurately adapted to the configuration of the virtual array (ULA linear trajectory or UCA circular trajectory), and to guarantee the accuracy and consistency of subsequent array guide vector construction, beam weighting, and other calculations, the specific forms of two types of vectors need to be defined according to the type of movement trajectory:

[0090] When the movement trajectory is a straight line trajectory adapted to ULA (used for one-dimensional angle measurement, estimating only azimuth or pitch angle), the dimensions of the three-dimensional spatial position vector and direction vector need to be simplified: on the one hand, the first... m Three-dimensional spatial position vector of each physical sampling location ( ), m =1,2..., MOnly the coordinate changes in the angular dimensions are retained, while the coordinates in non-angular dimensions remain fixed (e.g., when measuring azimuth along the x-axis). , For fixed values, only (This varies with the sampling position), ensuring the virtual array is linearly distributed; on the other hand, the direction vector of the search grid point only contains information in a single angular dimension, so if the azimuth angle is estimated, it would be... ( (For the azimuth of the search grid points), if the elevation angle is estimated, then it is... ( (For the pitch angle of the search grid points), the z-axis component is set to 0 to match the one-dimensional angle measurement characteristics of the linear array and avoid interference from irrelevant dimensions in the phase calculation.

[0091] When the movement trajectory is a circular trajectory adapted to UCA (used for two-dimensional angle measurement, simultaneously estimating azimuth and pitch angles), the vector form needs to reflect the circular distribution characteristics and the two-dimensional angle measurement requirements: Firstly, the... m The three-dimensional spatial position vector of each physical sampling location is defined as follows: , ( ), m =1,2..., M Where r is the radius of the circular trajectory, For the first m The circumferential position angle of each virtual array element ensures... M The sampling positions are uniformly distributed on a circular trajectory in the xy plane, forming an omnidirectional virtual circular array; secondly, the direction vector of the search grid points is defined as... It also includes azimuth. and pitch angle The information is presented, and the three-dimensional characteristics of spatial direction are reflected through the combination of trigonometric functions, which is precisely adapted to the two-dimensional angle measurement requirements of UCA, providing core vector support for the subsequent simultaneous estimation of angles in two dimensions.

[0092] See Figure 4 , Figure 4 This application provides a flowchart of a method for constructing adjacent sum and difference beams. Accordingly, in step S104, adjacent sum and difference beams are constructed based on the initial estimated angle from the DOA and the preset scanning interval to obtain angle-measuring adjacent beams. This can be specifically implemented through steps S401-S404.

[0093] S401: Based on the preset scanning interval, construct the left adjacent beam direction vector and the right adjacent beam direction vector.

[0094] To accurately capture angle deviation information within the neighborhood of the initial DOA-estimated angle and provide a beam pointing basis for subsequent sum-difference beam construction focused on the target direction, it is necessary to construct left and right adjacent beam direction vectors based on a preset scanning interval. and Using the initial estimated angle obtained from DBF as the center, two beam pointing directions are formed by shifting to both sides through a preset small scanning interval (balancing deviation capture accuracy and computational load, usually much smaller than the DBF division step size). The left adjacent beam pointing direction corresponds to the initial estimated angle minus the scanning interval, and the right adjacent beam pointing direction corresponds to the initial estimated angle plus the scanning interval. These two adjacent pointing directions can accurately cover the "deviation range between the initial estimated angle and the true angle", ensuring that subsequent sum and difference beams can effectively extract deviation information and avoid accuracy loss due to beam pointing deviating from the deviation range. During construction, the vector form needs to be adapted according to the angle measurement dimension (one-dimensional / two-dimensional) and the movement trajectory type (ULA / UCA): one-dimensional angle measurement only shifts to a single dimension (azimuth or elevation), while the other dimension is fixed; two-dimensional angle measurement requires constructing corresponding left and right adjacent vectors for the azimuth and elevation dimensions respectively to ensure that deviations in both dimensions can be captured.

[0095] S402: Construct left adjacent beam steering vector and right adjacent beam steering vector based on the left adjacent beam direction vector and the right adjacent beam direction vector respectively, and use them as left beam weighting vector and right beam weighting vector.

[0096] To convert the angular information of the left and right adjacent beam direction vectors into a signal weighting basis that the virtual array can execute, and to achieve phase alignment and energy extraction of the pointing signals of adjacent beams, it is necessary to base the calculation on the left adjacent beam direction vector. and the right adjacent beam direction vector Construct left and right adjacent array guide vectors respectively. and And directly use it as the left beam weighting vector. and right beam weighting vector Among them, the array steering vector integrates the wireless signal wavelength λ, M Three-dimensional spatial position vector of each physical sampling location , ,..., Spatial information of adjacent beam direction vectors, its mathematical expression (or This allows for the assignment of phase weights to each physical sampling location, matching the direction of the corresponding adjacent beam. When a signal is incident from that adjacent beam direction, the received signals at each sampling location are weighted and superimposed in phase, enhancing signal energy. Simultaneously, this weighting vector is... M 3D complex column vector (i.e., belonging to) Perfectly adapted M The signal processing requirements of each physical sampling location provide the core foundation for the subsequent construction of sum and difference beam weights and signal weighted summation.

[0097] It should be noted that, , ;

[0098] ;

[0099] ;

[0100] in, The wavelength of the wireless signal. , ,..., for M A three-dimensional spatial position vector of each physical sampling location.

[0101] S403: Construct the differential beam weights based on the left beam weighting vector and the right beam weighting vector, and determine the beam weights based on the DBF output obtained during the DBF angle estimation process.

[0102] To achieve the core functions of "enhancing the target direction signal" and "extracting angle deviation information" respectively, and to provide a key beam weighting basis for subsequent monopulse angle measurement, the left beam weighting vector can be used as a basis. and right beam weighting vector Difference beam weights are constructed through simple vector addition and subtraction operations. Differential beam weighting uses a "left-right" approach. The construction method amplifies the differences between the left and right beam signals, accurately capturing the deviation between the "true angle and the preliminary estimated angle." Simultaneously, to adapt to one-dimensional / two-dimensional angle measurement requirements, the weighted vectors of both the left and right beams are split according to the angle dimension. Includes left beam weighting vector of azimuth angle and pitch angle left beam weighting vector , Includes right beam weighting vector of azimuth angle And pitch angle right beam weighting vector .

[0103] Beam weight It can be determined based on the DBF output z obtained during the DBF angle estimation process: , The DBF output z with the largest modulus Corresponding direction vector The first in z n element Therefore, the azimuth angle and beam Consistent with the pitch angle and beam ,Right now y.

[0104] For example, assuming that the DOA estimation range is divided into grids, a total of N =5 search grid points, the corresponding direction vectors are as follows: , , , , After DBF beam output calculation, the DBF output amplitudes corresponding to each grid point are obtained as follows: | |=0.8, | |=1.2, | |=2.5, | |=1.6, | | = 0.9, with the largest modulus being | |=2.5, therefore we take its corresponding direction vector. As the optimal direction vector and the beam weighting vector corresponding to that direction. Directly used as beam weight ,Right now: .

[0105] Furthermore, to adapt to one-dimensional / two-dimensional angle measurement requirements, the corresponding beam weights are also split according to the angle dimension: the beam weights Including azimuth and beam weight ( (and pitch angle and beam weight) ( ); the difference beam weight Including azimuth difference beam weights ( and pitch difference beam weights ( ).

[0106] S404: Construct azimuth difference beams and / or elevation difference beams based on the sum beam weights, the difference beam weights, and the complex signal matrix, respectively.

[0107] To transform the function of sum and difference beamweighting into a quantifiable signal output, and to achieve both "enhancing the target direction signal" and "extracting angle deviation information," thus laying the data foundation for subsequent single-pulse sum and difference beam ratio calculation of angle deviation, it is necessary to rely on sum and difference beamweighting. Difference beam weights and the received signal matrix Construct azimuth sum difference beams and / or elevation sum difference beams. This is achieved through a conjugate transpose weighted summation operation ( ), and perform adaptation operations between the sum / difference beam weights and the received signal matrix, and sum beam weights ( , This enables phase alignment and energy superposition of signals in the target direction (near the initially estimated angle), resulting in a higher output intensity and beam signal. , Differential beam weighting (); , It can amplify the difference between the left and right adjacent beam signals and output a difference beam signal that reflects the angle deviation. , ).

[0108] It should be noted that this process is performed independently according to the angle dimension. In a one-dimensional angle measurement scenario, only a sum and difference beam of a single dimension (azimuth or elevation) needs to be constructed. In a two-dimensional angle measurement scenario, sum and difference beams of two dimensions need to be constructed simultaneously to ensure that the angle information of each dimension can be extracted independently, avoid interference between dimensions, and the operation is uniform and has low complexity, which is suitable for the real-time requirements of digital signal processing.

[0109] Steps S401-S404 quickly construct the left and right adjacent beam vectors and weighted vectors through a preset scanning interval, and then obtain the sum and difference beam weights through simple addition and subtraction operations. The sum and difference beams in the azimuth and elevation dimensions can be efficiently formed without complex calculations. This not only accurately focuses the target signal and extracts the angle deviation information, but also adapts to one-dimensional / two-dimensional angle measurement requirements, greatly reducing the complexity and computation of beam construction. Moreover, the connection with the complex signal matrix is ​​simple, laying an efficient and reliable foundation for subsequent high-precision deviation calculation.

[0110] In one possible implementation, in order to ensure that the left and right adjacent beam direction vectors are accurately adapted to the trajectory configuration of the virtual array (ULA straight trajectory or UCA circular trajectory) and the angle measurement dimension requirements, and to ensure the targeted extraction of angle deviation information, the dimensions of adjacent beam direction vectors need to be split according to the type of movement trajectory: the vector dimensions are strictly matched with the angle measurement requirements, one-dimensional angle measurement retains only the adjacent offset of a single dimension, and two-dimensional angle measurement covers both-dimensional offsets at the same time, avoiding interference from irrelevant dimensions.

[0111] Specifically, when the movement trajectory is a straight trajectory adapted to ULA (used for one-dimensional angle measurement, estimating only the azimuth or elevation angle), the left and right adjacent beam direction vectors contain only the offset of a single angle measurement dimension: the left adjacent beam direction vector Including the azimuth angle and the left adjacent beam direction vector (Fix initial pitch angle) Azimuth angle according to scan interval (Left skew) or pitch angle with the left adjacent beam direction vector (Fixed initial azimuth angle) Pitch angle according to scan interval (Left deviation); Right adjacent beam direction vector Including the azimuth angle of the adjacent beam direction vector (Azimuth angle rightward deviation) or elevation angle rightward deviation of adjacent beam direction vector (Pitch angle right skew) ensures that the vector focuses on single-dimensional deviation capture.

[0112] When the movement trajectory is a circular trajectory adapted to UCA (used for two-dimensional angle measurement, simultaneously estimating azimuth and elevation angles), the left and right adjacent beam direction vectors need to cover the two-dimensional offset: the left adjacent beam direction vector... Simultaneously includes (Azimuth angle deflected to the left) and (Pitch angle left deflection); Right adjacent beam direction vector Simultaneously includes (Azimuth deflection to the right) (Pitch angle right deflection) ensures that the deviation information of azimuth and pitch angles can be extracted independently, meeting the requirements of two-dimensional angle measurement.

[0113] In one possible implementation, in order to perform high-precision correction and compensation on the initial DOA angle estimate and obtain the final angle result, it is necessary to calculate the azimuth angle estimation deviation separately. and / or pitch angle estimation bias .

[0114] Specifically, azimuth estimation bias The calculation is based on Implementation: First, calculate the azimuth angle monopulse sum-difference beam ratio. ,in It is an azimuth difference beam, This refers to the azimuth angle and beamwidth. This ratio directly reflects the deviation trend between the true azimuth angle and the preliminary estimated azimuth angle. A ratio of 0 indicates no deviation, a positive ratio indicates the true angle is biased to the right, and a negative ratio indicates it is biased to the left. The larger the absolute value, the greater the deviation. The azimuth angle single-pulse slope... The pre-calibrated and stored calibration value, with the slope as the calibration coefficient, can eliminate the interference of signal amplitude on the sum and difference beam ratio, and convert the dimensionless ratio into the actual angle deviation value (unit: degree), ensuring the accuracy of deviation calculation under different signal strengths.

[0115] Pitch angle estimation deviation The calculation logic is completely consistent with that of the azimuth angle and is dimension-independent; the formula is: First calculate the pitch angle, monopulse sum, and difference beam ratio. ( For pitch difference beamforming, The ratio of pitch angle and beam (where the pitch angle and beam are both present) reflects the deviation trend of the pitch angle; while the pitch angle single-pulse slope... These are the pre-calibrated and stored calibration values. Finally, the pitch angle estimation deviation is obtained by multiplying them, thus achieving accurate correction of the initial pitch angle.

[0116] The entire process is calculated independently in terms of azimuth and elevation angles, which avoids information interference during multi-dimensional angle measurement and allows for the targeted extraction of deviation characteristics in each dimension. Combined with the calibration of the single-pulse slope, this ensures the accuracy and robustness of the angle estimation deviation calculation, providing reliable data support for subsequent high-precision angle correction.

[0117] See Figure 5 , Figure 5 This is a schematic diagram of an antenna angle measurement device based on a virtual array, provided as an embodiment of this application. Figure 5 As shown, the antenna angle measuring device based on a virtual array includes:

[0118] The selection unit 501, when performing the angle measurement task of the target antenna for the first time, is used to determine the movement trajectory of the target antenna according to the application angle measurement requirements of the target antenna, and uniformly select along the movement trajectory. M Each physical sampling position is recorded, along with its three-dimensional spatial coordinates. The movement trajectory is either a straight line trajectory adapted to ULA or a circular trajectory adapted to UCA. M It is a positive integer;

[0119] Virtual array antenna construction unit 502 is used to combine various three-dimensional spatial coordinates to construct the antenna array. M Each physical sampling location corresponds one-to-one with, which is equivalent to M Each virtual array antenna element, and the M The virtual array antenna elements are arranged to match the moving trajectory.

[0120] Complex signal matrix construction unit 503, used in the virtual array antenna M Wireless signals from a narrowband complex signal source to be located are collected at each physical sampling location. The wireless signals are then subjected to quadrature demodulation and analog-to-digital conversion using an orthogonal receiver to obtain a complex signal matrix. The wireless signals include signal amplitude, phase, and additive white Gaussian noise.

[0121] The preliminary estimation unit 504 is used to perform DBF angle estimation on the virtual array antenna in conjunction with the complex signal matrix to obtain the preliminary DOA estimated angle; the preliminary DOA estimated angle includes a preliminary azimuth angle and / or a preliminary elevation angle.

[0122] Angle-measuring adjacent beam construction unit 505 is used to construct adjacent sum and difference beams based on the preliminary angle estimated by the DOA and the preset scanning interval to obtain angle-measuring adjacent beams; the angle-measuring adjacent beams include azimuth sum and difference beams and / or elevation sum and difference beams; the azimuth sum and difference beams include azimuth sum beams and azimuth difference beams; the elevation sum and difference beams include elevation sum beams and elevation difference beams;

[0123] The correction and compensation unit 506 is used to solve the angle estimation deviation based on the adjacent beams of the angle measurement, and to use the angle estimation deviation to correct and compensate the initial estimated angle of the DOA to obtain the final angle estimation result; the angle estimation deviation includes the azimuth angle estimation deviation and / or the elevation angle estimation deviation.

[0124] In one possible implementation, the device further includes:

[0125] The setting unit is used to use the final angle estimation result of the previous time as the initial estimated angle of the current DOA when the target antenna is not performing the angle measurement task for the first time and moves repeatedly along the movement trajectory.

[0126] The non-first-time construction unit is used to construct adjacent sum and difference beams based on the initial estimated angle of this DOA and the preset scanning interval to obtain the adjacent beams for this angle measurement.

[0127] The non-first-time solution unit is used to solve the angle estimation deviation based on the adjacent beams of the current angle measurement.

[0128] The non-first-time correction and compensation unit is used to correct and compensate the initial estimated angle of the current DOA using the current angle estimation deviation to obtain the final angle estimation result for the current time; the current angle estimation deviation includes the current azimuth angle estimation deviation and / or the current pitch angle estimation deviation.

[0129] In one possible implementation, the complex signal matrix is:

[0130] ;

[0131] Where y is the complex signal matrix. , express M A complex field of 1 × 1 dimension; The first in m element Indicates the first m The signal received at each physical sampling location; It is additive white Gaussian noise. The first in m noise component It follows a mean of 0 and a variance of . Gaussian distribution; For any narrowband complex signal source pair to be located M The transmitted signal vector emitted from each location, , The first in m element , A The signal amplitude; For the target antenna in the 1st m The three-dimensional spatial position vector of the i-th physical sampling location is given by the i-th physical sampling location. m The three-dimensional spatial coordinates of each physical sampling location; Angle of arrival of the wireless signal directional vector, , The azimuth angle of the wireless signal; The elevation angle of the wireless signal is denoted as .

[0132] In one possible implementation, the preliminary estimation unit 504 is specifically used for:

[0133] The DOA estimation range of the target antenna is determined based on the installation attitude of the target antenna and the spatial distribution range of the narrowband signal source, and the division step size is determined.

[0134] The DOA estimation range of the target antenna is determined based on the installation attitude of the target antenna and the spatial distribution range of the narrowband signal source, and the division step size is determined.

[0135] The DOA estimation range is divided into grids based on the specified partitioning step size to obtain... N There are several search grids, and the direction vectors corresponding to the search grid points of each grid are obtained. ; n =1,2,… N , N It is a positive integer;

[0136] Based on the direction vectors corresponding to the search grid points of each search grid. Construct array guide vectors for each search grid As the first beam weighting vector ; , , The wavelength of the wireless signal. , ,..., for M A three-dimensional spatial position vector of each physical sampling location;

[0137] Using the first beam weighting vector The DBF output z is obtained by weighted summation of the wireless signals received at each physical sampling location in the received signal; The first in z n element ;

[0138] Based on the DBF output Determine the spatial spectrum ξ, and perform a spectral peak search on the spatial spectrum ξ to obtain the preliminary estimated angle of the DOA; The first in ξ n element .

[0139] In one possible implementation, when the movement trajectory is a straight-line trajectory adapted to ULA, the first... m Three-dimensional spatial position vector of each physical sampling location , ;No. n The direction vector corresponding to each search grid point or ,in, and These are the azimuth and elevation angles of the search grid points, respectively.

[0140] When the moving trajectory is a circular trajectory adapted to a uniform circular array (UCA), the first... m Three-dimensional spatial position vector of each physical sampling location , ;No. n The direction vector corresponding to each search grid point , For the first n The first search grid point k The azimuth of the road; For the first n The first search grid point The pitch angle of the road;

[0141] in, , It is a three-dimensional real Euclidean space; For the first m The position angle of each virtual array antenna element on the circumference r Let be the radius of the circular trajectory.

[0142] In one possible implementation, the angle-measuring adjacent beamforming unit 505 is specifically used for:

[0143] Based on the preset scanning interval, the left adjacent beam direction vector is constructed. and the right adjacent beam direction vector ;

[0144] Based on the left adjacent beam direction vector and the right adjacent beam direction vector Construct left adjacent beam steering vectors respectively and the right adjacent beam steering vector As the left beam weighting vector and right beam weighting vector ; , The wavelength of the wireless signal. , ,..., for M A three-dimensional spatial position vector of each physical sampling location; ;

[0145] Based on the left beam weighting vector and the right beam weighting vector Constructing difference beam weights Based on the DBF output z obtained during the DBF angle estimation process, the beam weights are determined. ; , The DBF output z with the largest modulus Corresponding direction vector The first in z n element ; ;

[0146] Based on the aforementioned beam weights The difference beam weight The complex signal matrices respectively construct azimuth and difference beams and / or elevation and difference beams;

[0147] Among them, the difference beam weight Including azimuth difference beam weights and pitch difference beam weights The azimuth angle and beamwidth are consistent with the elevation angle and beamwidth. y; the azimuth difference beam is The elevation difference beam is y is the complex signal matrix; the left beam weighting vector Including the left beam weighting vector of the azimuth angle and pitch angle left beam weighting vector The right beam weighting vector Including the right beam weighting vector of the azimuth angle And pitch angle right beam weighting vector , , .

[0148] In one possible implementation, when the movement trajectory is a straight trajectory adapted to ULA, the left adjacent beam direction vector Including the azimuth angle and the left adjacent beam direction vector Or the pitch angle and the left adjacent beam direction vector The right adjacent beam direction vector Including the azimuth angle of the adjacent beam direction vector Or the right adjacent beam direction vector of the elevation angle ;

[0149] When the movement trajectory is a circular trajectory of a uniform circular array (UCA), the left adjacent beam direction vector Including the azimuth angle and the left adjacent beam direction vector and the left adjacent beam direction vector of the elevation angle The right adjacent beam direction vector Including the azimuth angle of the adjacent beam direction vector And the right adjacent beam direction vector of the elevation angle ;

[0150] in, , , , , and These refer to the preliminary azimuth angle and the preliminary elevation angle, respectively. and All of these are the preset scanning intervals.

[0151] In one possible implementation, the angle estimation deviation includes azimuth estimation deviation and / or pitch estimation deviation;

[0152] The formula for the azimuth angle estimation deviation is as follows:

[0153] , The azimuth angle is estimated as a deviation. The single-pulse sum and difference beam ratio at the azimuth angle. , For azimuth difference beam, For azimuth and beam; The slope of the single pulse is the azimuth angle;

[0154] The formula for the pitch angle estimation deviation is as follows:

[0155] ， Estimate the deviation for the pitch angle. The single-pulse sum-difference beam ratio for elevation angle, ， For pitch difference beam, For pitch angle and beam; The slope of the single pulse represents the pitch angle.

[0156] In addition, this application embodiment also provides an antenna angle measurement device based on a virtual array, including: a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, it implements the antenna angle measurement method based on a virtual array as described above.

[0157] The embodiments of this application utilize the construction of virtual arrays to reduce reliance on large-scale physical antenna arrays, significantly reducing hardware costs and system complexity. Furthermore, by equating physical sampling positions with virtual array antenna elements, it can flexibly adapt to different angle measurement requirements.

[0158] The foregoing has provided a detailed description of an antenna angle measurement method, apparatus, and device based on a virtual array provided in this application. The various embodiments in the specification are described in a progressive manner, with each embodiment focusing on its differences from other embodiments. Similar or identical parts between embodiments can be referred to interchangeably. For the apparatus disclosed in the embodiments, since it corresponds to the method disclosed in the embodiments, the description is relatively simple; relevant parts can be referred to in the method section. It should be noted that those skilled in the art can make several improvements and modifications to this application without departing from the principles of this application, and these improvements and modifications also fall within the protection scope of the claims of this application.

[0159] It should also be noted that, in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus.

Claims

1. An antenna angle measurement method based on a virtual array, characterized in that, The method includes: When performing the angle measurement task of the target antenna for the first time, the movement trajectory of the target antenna is determined according to the application angle measurement requirements of the target antenna, and uniform selections are made on the movement trajectory. M Each physical sampling position is recorded, and the three-dimensional spatial coordinates of each physical sampling position are recorded; the movement trajectory is a straight trajectory adapted to a uniform linear array ULA, or a circular trajectory adapted to a uniform circular array UCA. M It is a positive integer; Combining the various three-dimensional spatial coordinates to the M Each physical sampling location corresponds one-to-one with, which is equivalent to M Each virtual array antenna element, and the M The virtual array antenna elements are arranged to match the moving trajectory. The virtual array antenna M Wireless signals from a narrowband signal source to be located are collected at each physical sampling location. The wireless signals are then subjected to quadrature demodulation and analog-to-digital conversion using an orthogonal receiver to obtain a complex signal matrix. The wireless signals include signal amplitude, phase, and additive white Gaussian noise. The virtual array antenna is subjected to digital beamforming (DBF) angle estimation using the complex signal matrix to obtain a preliminary estimated Direction of Arrival (DOA) angle. Based on the preliminary DOA angle and a preset scanning interval, adjacent sum and difference beams are constructed to obtain angle-measuring adjacent beams. The preliminary DOA angle includes a preliminary azimuth angle and / or a preliminary elevation angle. The angle-measuring adjacent beams include azimuth sum and difference beams and / or elevation sum and difference beams. The azimuth sum and difference beams include an azimuth sum beam and an azimuth difference beam. The elevation sum and difference beams include an elevation sum beam and an elevation difference beam. The angle estimation deviation is calculated based on the adjacent beams of the angle measurement, and the angle estimation deviation is used to correct and compensate the initial angle estimate of the DOA to obtain the final angle estimation result; the angle estimation deviation includes the azimuth angle estimation deviation and / or the elevation angle estimation deviation.

2. The method according to claim 1, characterized in that, The method further includes: When the target antenna is not performing the angle measurement task for the first time and moves repeatedly along the movement trajectory, the final angle estimation result of the previous time is used as the initial estimated angle of the current DOA. Based on the preliminary estimated angle of this DOA and the preset scanning interval, adjacent sum and difference beams are constructed to obtain the adjacent beams for this angle measurement. The angle estimation deviation for this angle measurement is calculated based on the adjacent beams of the current angle measurement. The initial DOA angle estimate is corrected and compensated using the current angle estimation deviation to obtain the final angle estimation result; the current angle estimation deviation includes the current azimuth angle estimation deviation and / or the current pitch angle estimation deviation.

3. The method according to claim 1, characterized in that, The complex signal matrix is: ； Where y is the complex signal matrix. , express M A complex field of 1 × 1 dimension; The first in m element Indicates the first m The signal received at each physical sampling location; It is additive white Gaussian noise. The first in m noise component It follows a mean of 0 and a variance of . Gaussian distribution; For any narrowband complex signal source pair to be located M The transmitted signal vector emitted from each location, , The first in m element , A For signal amplitude, λ The wavelength of the wireless signal; For the target antenna in the 1st m The three-dimensional spatial position vector of the i-th physical sampling location is given by the i-th physical sampling location. m Composed of the three-dimensional spatial coordinates of each physical sampling location; Angle of arrival of the wireless signal directional vector, , The azimuth angle of the wireless signal; The elevation angle of the wireless signal is denoted as .

4. The method according to claim 1, characterized in that, The preliminary estimated DOA angle is obtained by performing DBF angle estimation on the virtual array antenna using the complex signal matrix, including: The DOA estimation range of the target antenna is determined based on the installation attitude of the target antenna and the spatial distribution range of the narrowband signal source, and the division step size is determined. The DOA estimation range is divided into grids based on the specified partitioning step size to obtain... N There are several search grids, and the direction vectors corresponding to the search grid points of each grid are obtained. ; n =1,2,… N , N It is a positive integer; Based on the direction vectors corresponding to the search grid points of each search grid. Construct array guide vectors for each search grid As the first beam weighting vector ; , , The wavelength of the wireless signal. , ,..., for M A three-dimensional spatial position vector of each physical sampling location; Using the first beam weighting vector The DBF output z is obtained by weighted summation of the wireless signals received at each physical sampling location in the received signal; The first in z n element ; Based on the DBF output Determine the spatial spectrum ξ, and perform a spectral peak search on the spatial spectrum ξ to obtain the preliminary estimated angle of the DOA; The first in ξ n element .

5. The method according to claim 4, characterized in that, When the movement trajectory is a straight line trajectory adapted to ULA, the first m Three-dimensional spatial position vector of each physical sampling location , ;No. n The direction vector corresponding to each search grid point or ,in, and These are the azimuth and elevation angles of the search grid points, respectively. When the moving trajectory is a circular trajectory adapted to a uniform circular array (UCA), the first... m Three-dimensional spatial position vector of each physical sampling location , ;No. n The direction vector corresponding to each search grid point , For the first n The first search grid point k The azimuth of the road; For the first n The first search grid point The pitch angle of the road; in, , It is a three-dimensional real Euclidean space; For the first m The position angle of each virtual array antenna element on the circumference r Let be the radius of the circular trajectory.

6. The method according to claim 1, characterized in that, The step of constructing adjacent sum and difference beams based on the preliminary estimated angle of the DOA and the preset scanning interval to obtain angle-measuring adjacent beams includes: Based on the preset scanning interval, the left adjacent beam direction vector is constructed. and the right adjacent beam direction vector ; Based on the left adjacent beam direction vector and the right adjacent beam direction vector Construct left adjacent beam steering vectors respectively and the right adjacent beam steering vector As the left beam weighting vector and right beam weighting vector ; , The wavelength of the wireless signal. , ,..., for M A three-dimensional spatial position vector of each physical sampling location; ; Based on the left beam weighting vector and the right beam weighting vector Constructing difference beam weights Based on the DBF output z obtained during the DBF angle estimation process, the beam weights are determined. ; , The DBF output z with the largest modulus Corresponding direction vector The first in z n element ; ; Based on the aforementioned beam weights The difference beam weight The complex signal matrices respectively construct azimuth and difference beams and / or elevation and difference beams; Among them, the difference beam weight Including azimuth difference beam weights and pitch difference beam weights The azimuth angle and beamwidth are consistent with the elevation angle and beamwidth. y; the azimuth difference beam is The elevation difference beam is y is the complex signal matrix; the left beam weighting vector Including the left beam weighting vector of the azimuth angle and pitch angle left beam weighting vector The right beam weighting vector Including the right beam weighting vector of the azimuth angle And pitch angle right beam weighting vector , , .

7. The method according to claim 6, characterized in that, When the movement trajectory is a straight line trajectory adapted to ULA, the left adjacent beam direction vector Including the azimuth angle and the left adjacent beam direction vector Or the pitch angle and the left adjacent beam direction vector The right adjacent beam direction vector Including the azimuth angle of the adjacent beam direction vector Or the right adjacent beam direction vector of the elevation angle ; When the movement trajectory is a circular trajectory of a uniform circular array (UCA), the left adjacent beam direction vector Including the azimuth angle and the left adjacent beam direction vector and the left adjacent beam direction vector of the elevation angle The right adjacent beam direction vector Including the azimuth angle of the adjacent beam direction vector And the right adjacent beam direction vector of the elevation angle ; in, , , , , and These refer to the preliminary azimuth angle and the preliminary elevation angle, respectively. and All of these are the preset scanning intervals.

8. The method according to claim 1, characterized in that, The angle estimation deviation includes the azimuth angle estimation deviation and / or the pitch angle estimation deviation; The formula for the azimuth angle estimation deviation is as follows: , The azimuth angle is estimated as a deviation. The single-pulse sum and difference beam ratio at the azimuth angle. , For azimuth difference beam, For azimuth and beam; The slope of the single pulse is the azimuth angle; The formula for the pitch angle estimation deviation is as follows: ， Estimate the deviation for the pitch angle. The single-pulse sum-difference beam ratio for elevation angle, ， For pitch difference beam, For pitch angle and beam; The slope of the single pulse is the pitch angle.

9. An antenna angle measurement device based on a virtual array, characterized in that, The device includes: The selection unit, when performing the angle measurement task of the target antenna for the first time, is used to determine the movement trajectory of the target antenna according to the application angle measurement requirements of the target antenna, and uniformly selects along the movement trajectory. M Each physical sampling position is recorded, along with its three-dimensional spatial coordinates. The movement trajectory is either a straight line trajectory adapted to ULA or a circular trajectory adapted to UCA. M It is a positive integer; Virtual array antenna construction unit, used to combine various three-dimensional spatial coordinates to... M Each physical sampling location corresponds one-to-one with, which is equivalent to M Each virtual array antenna element, and the M The virtual array antenna elements are arranged to match the moving trajectory. Complex signal matrix construction unit, used in the virtual array antenna M Wireless signals from a narrowband signal source to be located are collected at each physical sampling location. The wireless signals are then subjected to quadrature demodulation and analog-to-digital conversion using an orthogonal receiver to obtain a complex signal matrix. The wireless signals include signal amplitude, phase, and additive white Gaussian noise. The preliminary estimation unit is used to perform DBF angle estimation on the virtual array antenna in conjunction with the complex signal matrix to obtain the preliminary DOA estimated angle; the preliminary DOA estimated angle includes a preliminary azimuth angle and / or a preliminary elevation angle. An angle-measuring adjacent beam construction unit is used to construct adjacent sum and difference beams based on the initial angle estimate of the DOA and the preset scanning interval to obtain angle-measuring adjacent beams; the angle-measuring adjacent beams include azimuth sum and difference beams and / or elevation sum and difference beams; the azimuth sum and difference beams include azimuth sum beams and azimuth difference beams; the elevation sum and difference beams include elevation sum beams and elevation difference beams; The correction and compensation unit is used to solve the angle estimation deviation based on the adjacent beams of the angle measurement, and to use the angle estimation deviation to correct and compensate the initial estimated angle of the DOA to obtain the final angle estimation result; the angle estimation deviation includes the azimuth angle estimation deviation and / or the elevation angle estimation deviation.

10. An antenna angle measurement device based on a virtual array, characterized in that, include: A memory, a processor, and a computer program stored in the memory and running on the processor, wherein the processor, when executing the computer program, implements the antenna angle measurement method based on a virtual array as described in any one of claims 1-8.

Images